Methodology for Predicting Future Stock Price

Prediction of future stock prices

The prediction of future prices from a time series of past raw stock prices has been recognized to be a very difficult problem, such a time series may be noisy and non-stationary, and many factors leading to price fluctuations cannot be captured precisely or may be too difficult to be modeled.

Time Series are popular in many application areas such as Daily fluctuations of a stock market, scientific experiments, Medical treatments, Weather conditions and so on.

- Trend Analysis

- Similarity Search

- Sequential Pattern Mining

- Periodicity Analysis

Methodology Application

- Stock Market data

- The main interest is to predict the timing of future events, ie. The points at which the stock will change the direction of its slope

- Stock prices are noisy and influenced daily by many factors

Time series prediction problem is the prediction of future values based on the previous values and the current value of the time series. The previous values and the current value of the time series are used as inputs for the prediction model.

The long-term prediction is typically faced with growing uncertainties arising from various sources. For instance, the accumulation of errors and the lack of information make the prediction more difficult.

The Time Series Data Mining framework is applied to the prediction of financial time series. It can successfully characterize and predict complex, non-periodic, irregular and chaotic time series.

A time series is “a sequence of observed data, usually ordered in time”.

Time series analysis is fundamental to engineering, scientific, and business endeavors, such as the prediction of welding droplet releases and stock market price fluctuations.

The Autogressive Integrated Moving Average techniques provide a comprehensive approach for analyzing stationary time series whose residuals are normal and independent. For real0world time series such as stock...